Fix bug: Enable multi-GPU prediction after activating . 1个月前 README Apache-2.0 SkyReels V1: Human-Centric Video Foundation Model 🔥🔥🔥 News!! 🎥 Demos 📑 TODO List 🌟 Overview 🔑 Key Features 1. Self-Developed Data Cleaning and Annotation Pipeline ...
return comfy.model_base.ModelType.V_PREDICTION class HYDiT_Model(comfy.model_base.BaseModel): def __init__(self, *args, **kwargs): super().__init__(*args, **kwargs) def extra_conds(self, **kwargs): out = super().extra_conds(**kwargs) addit_embeds = kwargs['cross...
( "--enable_zero_snr", action="store_true", default=False, help="whether use zero_snr in scheduler, include v_prediction、trailing, etc , default=`False`", ) # moore animateanyone end parser.add_argument( "--record_mid_video_noises", action="store_true", default=False, help="...
For different types of input condition, such as reference image and target audio, we provide parameters for adjusting the role played by that condition information in the model prediction. We refer to these two parameters as reference_attention_weight and audio_attention_weight. Different parameters ...
We implement GenAgent on the ComfyUI platform and propose a new benchmark, OpenComfy. The results demonstrate that GenAgent outperforms baseline approaches in both run-level and task-level evaluations, showing its capability to generate complex workflows with superior effectiveness and stability. 2nd...
- [ ] ComfyUI ## 🌟 Overview SkyReels V1 is the first and most advanced open-source human-centric video foundation model. By fine-tuning HunyuanVideo on O(10M) high-quality film and television clips, SkyReels V1 offers three key advantages: @@ -19,11 +56,6 @@ SkyReels V1 is...
For different types of input condition, such as reference image and target audio, we provide parameters for adjusting the role played by that condition information in the model prediction. We refer to these two parameters asreference_attention_weightandaudio_attention_weight. Different parameters can ...
- Update the I2V model and related training prediction code to version 1.5. [2024.12.16] - Use reward backpropagation to train Lora and optimize the video, aligning it better with human preferences, detailes in [here](scripts/README_TRAIN_REWARD.md). A new version of the control model ...
ComfyUI integration Diffusers integration Diffusers + Multi-GPU Inference Wan2.1 Image-to-Video Multi-GPU Inference code of the 14B model Checkpoints of the 14B model Gradio demo ComfyUI integration Diffusers integration Diffusers + Multi-GPU Inference Quickstart Installation Clone the repo: git clone...
ComfyUI 🌟 Overview SkyReels V1 is the first and most advanced open-source human-centric video foundation model. By fine-tuningHunyuanVideoon O(10M) high-quality film and television clips, SkyReels V1 offers three key advantages: Open-Source Leadership: Our Text-to-Video model achieves stat...